Hierarchical Indian Buffet Neural Networks for Bayesian Continual Learning
This addresses the challenge of overfitting and underfitting in continual learning for AI systems, though it is incremental as it builds on existing Bayesian and IBP methods.
The paper tackles the problem of resource allocation in continual learning by introducing a hierarchical Indian Buffet process prior over Bayesian neural networks, allowing automatic adjustment of network complexity to handle new tasks, and shows empirically that it offers a competitive edge over existing methods.
We place an Indian Buffet process (IBP) prior over the structure of a Bayesian Neural Network (BNN), thus allowing the complexity of the BNN to increase and decrease automatically. We further extend this model such that the prior on the structure of each hidden layer is shared globally across all layers, using a Hierarchical-IBP (H-IBP). We apply this model to the problem of resource allocation in Continual Learning (CL) where new tasks occur and the network requires extra resources. Our model uses online variational inference with reparameterisation of the Bernoulli and Beta distributions, which constitute the IBP and H-IBP priors. As we automatically learn the number of weights in each layer of the BNN, overfitting and underfitting problems are largely overcome. We show empirically that our approach offers a competitive edge over existing methods in CL.